Collation and distribution of information and data on experimental and clinical studies related to medical treatment is a valuable service provided by centers such as the National Cancer Institute. This information may be in the form of old program records or unformatted, handwritten notes which are awkward to store, retrieve, and transmit. A novel approach is proposed for the conversion of these records to computerized, digital formats. The proposed technique uses software control of input parameters such as source illumination and sharpness of focus to augment the optical character recognition (OCR) process. This approach is particularly adapted to the reading of old records which may exhibit irregularities in print contrast and legibility. In addition, this approach requires less computing power than conventional OCR techniques and can be implemented on a microcomputer. The feasibility of this approach will be determined in Phase I, and supporting data for the Phase II prototype design will be collected. The commercial potential of our proposed optical reader is considerable in that relatively inexpensive devices could be developed. Thus, optical reading technology would be available to local information distribution centers such as libraries and schools, as well as to handicapped individuals.